All publications and patent applications herein are incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.
The following description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed inventions, or that any publication specifically or implicitly referenced is prior art.
Plants grown for commercial agricultural purposes are nearly always planted as uniform monocultures; that is, single varieties of a given crop are mass-produced by vegetative propagation or by seed and are planted on a very large scale. When a pathogen or pest arrives that can overcome the natural disease or pest resistance of a given variety, severe economic losses can occur because of the practice of monoculture, sometimes involving loss of the entire crop in a given area. Control of diseases and pests using massive applications of agricultural chemicals is expensive, environmentally unsound and often impossible. For example, citrus canker disease, caused by a quarantined Gram-negative bacterial pathogen, Xanthomonas citri, has spread uncontrollably throughout Florida. As a second example, the Gram-negative bacterial pathogen Ca. Liberibacter asiaticus is a USDA Select Agent (potential bioterrorist agent) that was introduced into Florida in 2005 and has spread uncontrollably throughout Florida. This pathogen threatens world citrus production. As a third example, the Gram negative bacterial pathogen Ralstoma solanacearum Race 3 Biovar 2 has been introduced into the U.S. numerous times and is such a serious threat to U.S. potato production that it is also a listed USDA Select Agent. This pathogen has been introduced into the U.S. by infecting geranium plants, but asymptomatically, so that detection of the pathogen is delayed.
As a fourth and final example, serious human illness and even deaths have been reported due to the Gram-negative bacterium Escherichia coli, which is capable of internally infecting—not just contaminating—certain crop plants such as spinach, alfalfa sprouts and mung bean sprouts. Several outbreaks of Salmonella and E. coli O157:H7 associated with organically grown sprouts and mesclun lettuce have been reported (Doyle, M. P. 2000. Nutrition 16: 647-9). According to the FDA in its web report of the 2006 outbreak of E. coli in contaminated spinach “To date, 204 cases of illness due to E. coil O157:H7 infection have been reported to the CDC including 31 cases involving a type of kidney failure called Hemolytic Uremic Syndrome (HUS), 104 hospitalizations, and three deaths. The first death was an elderly woman in Wisconsin; the second death, a two-year-old in Idaho; and the third death, an elderly woman in Nebraska.” Conventional plant breeding to control such diseases of plants or food-borne contamination has proven to be impossible. There is therefore an urgent and pressing need for gene engineering techniques to provide plants, including carrier plants such as geraniums, with disease and pest resistance against diseases and pests that they naturally are susceptible to, or tolerant of.
A wide variety of antibacterial and antifungal proteins have been identified and their genes isolated from both animals and plants. Because of the major differences in the structures of fungal, Gram-positive bacterial and Gram-negative bacterial cell walls, many of these proteins attack only fungi or Gram positive bacteria, which have cell walls that are exposed directly to the environment. Gram negative bacteria do not have cell walls that are exposed directly to the environment. Instead, their cell walls are enveloped and protected by a unique outer membrane structure, the lipopolysaccharide (LPS) barrier, which provides a very effective additional barrier to protect their cell walls against most eukaryotic defenses, particularly plant defenses. The great majority of the pathogens listed by the USDA as Select Agents are bacterial pathogens, and all of these are Gram negative.
The LPS provides an effective defense to Gram negative bacteria against externally produced enzymes that can effectively degrade the bacterial cell wall (also called the murein layer), including the relatively thick but exposed cell walls of Gram-positive bacteria and fungi. For example, lysozymes are antimicrobial agents found in mammalian cells, insects, plants, bacteria and viruses that break bacterial and fungal cell walls, specifically cleaving bonds between the amino sugars of the recurring muropeptides (C-1 of N-acetylmuramic acid and C-4 of N-acetylglucosamine of microbial cell walls (Ibrahim et al. 2001 and references therein). Some lysozymes also are pleiotropically lytic proteins, meaning they are active in killing Gram-negative and Gram-positive bacteria, but this activity is not due to the enzymatic action of lysozyme, but specifically due to a short, linear peptide fragment that is a degradation product of some lysozymes; it is the linear degradation product of the lysozyme that penetrates the LPS barrier and the cell wall (but without harming either), reaching the inner membrane and permeabilizing the inner membrane, resulting in lysis (During et al, 1999; Ibrahim et al. 2001). However, this linear peptide activity does not work well in plants (see below).
Those antimicrobial proteins demonstrated to kill Gram-negative bacteria are mostly small peptides (proteins of less than 50 amino acids in length) that are amphipathic and positively charged, so that they are attracted to the negatively charged Gram negative outer membrane, are small enough to penetrate can penetrate the exposed LPS, and also the relatively thin Gram negative cell wall. These peptides usually act to permeabilize the inner membrane, directly causing cell death. During the last two decades, over 500 antimicrobial peptides have been discovered in viruses, insects, plants and animals (Jaynes et al, 1987; Mitra and Zhang, 1994; Broekaert et al. 1997; Nakajima et al, 1997; Vunnam et al, 1997). The best described of these are peptides having broad spectrum activity in the source organism and in artificial media against viruses, bacteria, fungi, parasites and even tumor cells (Hancock and Lehrer, 1998).
The largest described group by far of these antimicrobial peptides are linear (eg., cecropins, attacins and magainins). However, linear peptides are not found naturally in plants and most linear peptides are rapidly degraded by plant proteases. For example, cecropin B is rapidly degraded when incubated with intercellular plant fluid, with a half-life ranging from about three minutes in potato to about 25 hours in rice (Owens & Heutte, 1997). Transgenic tobacco plants expressing cecropins have only slightly increased resistance to (Gram-negative) Pseudomanas syringae pv. tabaci, the cause of tobacco wildfire (Huang et al 1997). Synthetic cecropin analogs Shiva-1 and SB-37, expressed from transgenes in potato plants, only slightly reduced bacterial infection caused by (Gram-negative) Erwinia carotovora (Arce et al 1999). Transgenic apple expressing the SB-37 peptide showed only slightly increased resistance to (Gram-negative) E. amylovora in field tests (Norelli et al 1998). Similarly, transgenic potatoes expressing attacin showed resistance to bacterial infection by E. carotovora (Arce et al 1999) and transgenic pear and apple expressing attacin genes have also shown slightly enhanced resistance to E. amylovora (Norelli et al 1994; Reynoird et al 1999). Attacin E was also found to be rapidly degraded by plants (Ko et al 2000). Transgenic tobacco plants expressing a synthetic magainin analog that had been modified to be less sensitive to extracellular plant proteases were only slightly resistant to the bacterial pathogen E. carotovora (Li et al 2001).
The disulfide-linked peptides (e.g. defensins, prophenins and thaumatins) show more promise of stability when expressed in plants, but resistance has either been weak, not demonstrated, or cytotoxicity issues have emerged. Hen egg-white lysozyme genes (with lytic ability) have been used to confer weak Gram-negative bacterial disease resistance to transgenic tobacco plants (Trudel et al 1995; Kato et al 1998). Bacteriophage T4 lysozyme has also been reported to slightly enhance resistance in transgenic potato against E. carotovora (During et al 1993; Ahrenholz et al., 2000) and in transgenic apple plants against E. amylovora (Ko 1999). However, as mentioned previously, the action of lysozyme against Gram-negative bacteria is specifically due to a short lytic peptide fragment (Ibrahim et al. 2001) that is presumably sensitive to protease. Thaumatins exhibit the widest range of antimicrobial activity so far characterized, but also exhibit potent cytotoxic effects on eukaryotic cells (Taguchi et al 2000). Defensins, produced by plants, mammals and insects, are characterized by complex β-sheet structures with several disulfide bonds that bind and disrupt microbial plasma membranes. A plant defensin from alfalfa gave robust resistance to a fungal pathogen (Guo et al 2000) and defensins from spinach were active in vitro against Gram positive and Gram negative bacteria (Segura et al. 1998). However, human illnesses have resulted from both alfalfa and spinach infected with enteric bacteria; evidently these defensins are either not triggered by these bacteria or they are ineffective against these bacteria. More effective antibacterial agents are urgently needed to protect crop plants.
Nonenzymatic, antimicrobial peptides are abundant in nature but of limited value in transgenic plants, primarily due to degradation by plant proteases. In addition, some Gram-negative bacteria are resistant to antimicrobial peptides even in culture media, due to variations in the chemical structure of the LPS (Gutsmann et al., 2005). This may help explain why plant pathogenic bacteria can overcome host plant defensins. To date, no antimicrobial peptide has proved more than marginally effective against Gram-negative bacteria when expressed in plants. More efficacious methods to control plant disease are urgently needed.
By contrast with bacterial pathogens of animals, the vast majority of bacterial pathogens of plants are all Gram negative. As mentioned above, the distinguishing feature of Gram-negative bacteria is the presence of the LPS, which forms an outer membrane that completely surrounds the cell wall. Mutations affecting the structure of the LPS of a (Gram-negative) bacterial plant pathogen of citrus caused the pathogen to die out very quickly on citrus, but not on bean (Kingsley et al., 1993), indicating the importance of the LPS structure in evading specific plant phytochemical defenses. In addition, mutations affecting multidrug efflux in Gram-negative bacteria cause the bacteria to die out rapidly in plants, highlighting the role of low molecular weight plant defense compounds (phytoalexins) in plant defense, and further indicating the importance of the intact LPS of Gram-negative in resisting plant defense compounds (Reddy et al., 2007). Multidrug efflux requires an intact LPS for function.
Animals have a unique set of innate defenses against microbial invasion that is independent of prior exposure to pathogens (Hoffman et al., 1999). Among these are the lytic peptides discussed above, and also the neutrophil, a white blood cell that is part of the innate immune system. Neutrophils produce a variety of protein and peptide antibiotics that kill microorganisms. Among these is the bactericidal/permeability increasing (BPI) protein, which is a potent antimicrobial protein that is primarily active towards Gram negative bacteria (Levy, 2000). BPI is not toxic to Gram positive bacteria, fungi or animal cells, but rather attacks the LPS layer of Gram negative cells, disrupting its structure, and eventually attacking the inner membrane and causing lysis (Mannion et al., 1990). A hallmark of BPI proteins is their strongly cationic, lysine rich nature and their opsonic or immune system activation ability (Levy et al., 2003). Members of the BPI protein family include lipopolysaccharide binding protein (LBP), lung specific X protein (LUNX), palate, lung and nasal epithelial clone (PLUNC) and parotid secretory protein (PSP), many of which have been identified by bioinformatics techniques with up to 43% identity between family members (Wheeler et al. 2003). There are numerous patents covering use of BPI and certain smaller peptide derivatives (for example, U.S. Pat. No. 5,830,860 and U.S. Pat. No. 5,948,408).
Antimicrobial Bacteriophage Proteins.
All bacteriophages must escape from bacterial host cells, either by extrusion from the host cell, as with filamentous phages, or by host cell lysis from within. Host cell lysis from within requires two events: ability to penetrate the inner membrane of both gram negative and gram positive bacteria, and ability to depolymerize the murein layer, which is relatively thick in gram positive cell walls.
Bacteriophage penetration of, and egress through, the inner membrane is accomplished in many, but evidently not all, phage by use of small membrane-localized proteins called “holins” that appear to accumulate in the bacterial inner membrane until reaching a specific concentration, at which time they are thought to self-assemble to permeabilize the inner membrane (Grundling et al., 2001; Wang et al. 2000; Young et al., 2000). The terms “holin” and “holin-like” are not biochemically or even functionally accurate terms, but instead as used herein refer to any phage protein with at least one transmembrane domain that is capable of permeabilizing the inner membrane, thereby allowing molecules other than holins that are normally sequestered in the cyctoplasm by the inner membrane, including proteins such as endolysins, to breach or penetrate the inner membrane to reach the cell wall. The biochemical function(s) of holins is speculative; most, if not all of the curent knowledge on holins is based on the λ phage S protein (Haro et al. 2003).
Holins are encoded by genes in at least 35 different families, having at least one transmembrane domain and classified into three topological classes (classes I, II, and III, with three, two and one transmembrane domains [TMD], respectively), all with no detected orthologous relationships (Grundling et al., 2001). At least two holins are known to be hemolytic and this hemolytic function has been hypothesized to play a role in the pathogenesis of certain bacteria towards insects and nematodes (Brillard et al., 2003). Only a few have been partially characterized in terms of in vivo function, leading to at least two very different theories of how they may function. The most widely accepted theory is that holins function to form oligomeric membrane pores (Graschopf & Blasi, 1999; Young et al., 2000).
Depolymerization of the murein layer is accomplished by lytic enzymes called endolysins. There are at least three functionally distinct classes of endolysins: 1) glucosaminidases (lysozymes) that attack the glycosidic linkages between the amino sugars of the peptidoglycan; 2) amidases that attack the N-acetylmuramoyl-L-alanine amide linkage between the glycan strand and the cross-linking peptide, and 3) endopeptidases that attack the interpeptide bridge linkages (Sheehan et al., 1997). Endolysins are synthesized without an export signal sequence that would permit them access to the peptidoglycan (murein) layer, and they therefore usually accumulate in the cytoplasm of phage infected bacteria until they are released by the activity of holins (Young and Blasi, 1995).
Lysozymes have been suggested as useful antibiotics that can be used as external agents against both Gram-positive and Gram-negative bacteria because at least some of them are multifunctional (During et al., 1999). This dual functionality is based on the finding that both phage T4 and hen egg white lysozyme have both glucosaminidase activity as well as amphipathic helical stretches that allow them to penetrate and disrupt bacterial, fungal and plant membranes (During et al., 1999). The microbicidal activity of lysozymes can be affected by C-terminal additions; additions of hydrophobic amino acids decreased activity against Gram positive bacteria, but increased activity against Gram negative E. coli (Arima et al., 1997; Ito et al., 1997). Additions of histidine, a hydrophilic amino acid, to T4 lysozyme doubled its antimicrobial activity against Gram-positive and Gram-negative bacteria (During et al., 1999).
The nonenzymatic, microbicidal function of lysozymes appeared to be due to amphipathic C-terminal domains that could be mimicked by small synthetic peptides modeled after the C-terminal lysozyme domains (During et al., 1999). As described above, transgenic plants have been created that express lysozymes and give some resistance to certain plant pathogens. Since most endolysins accumulate to high titers within the bacterial cell without causing lysis, endolysins other than certain lysozymes such as T4 would not be expected to attack Gram-negative bacteria if externally applied, since Gram-negative bacteria are surrounded with an outer membrane comprised of LPS and a lipid bilayer that would protect its murein layer from enzymatic attack just as effectively as its inner membrane does.
Attempts have been made to treat bacterial diseases of both animals and plants by use of intact bacteriophage. All of these attempts have severe limitations in their utility. For examples, U.S. Pat. No. 5,688,501 discloses a method for treating an infectious disease of animals using intact bacteriophage specific for the bacterial causal agent of that disease. U.S. Pat. No. 4,957,686 discloses a method for preventing dental caries by using intact bacteriophage specific for the bacterial causal agent of dental caries. Flaherty et al. (2000) describe a method for treating an infectious disease of plants using intact bacteriophage specific for the bacterial causal agent of that disease. In all these cases and in similar cases using intact bacteriophage, the bacteriophage must attach to the bacterial host, and that attachment is highly host specific, limiting the utility of the phage to specific bacterial host species, and sometimes specific bacterial host strains. In addition, for attachment to occur, the bacteria must be in the right growth phase, and the phage must be able to gain access to the bacteria, which are often buried deep within tissues of either animals or plants, or shielded by bacterial biofilms, formed in part by the secretion of bacterial extracellular polysaccharides (EPS).
Attempts have been made to treat Erwinia amylovora bacterial infections of pear and apple trees through the use of transgenic plants expressing an extracellular polysaccharide (EPS) degrading enzyme, EPS-depolymerase, derived from an E. amylovora phage. However, the level of resistance achieved was weak, at best, and the phage EPS-depolymerase was very specific for the EPS from E. amylovora. More efficacious, and more generally applicable, strategies are clearly needed.
Attempts have been made to treat gram-positive bacterial diseases of animals, but not plants, by use of lytic enzyme preparations extracted from bacteriophage infected bacteria or from bacteria expressing bacteriophage genes. These, too, have serious limitations. For example, U.S. Pat. No. 5,985,271 discloses a method of treating an animal disease caused by a specific gram positive bacterium, Streptococcus, by use of a crude specific endolysin preparation. Similarly, U.S. Pat. No. 6,017,528 discloses a method of preventing and treating Streptococcus infection of animals by use of a crude specific endolysin preparation. Similarly, WO 01/90331 and US 2002/0058027 disclose methods of preventing and treating Streptococcus infection of animals by use of a purified preparation consisting of a specific endolysin. In all of these cases, the enzyme preparations must be purified, buffered, prepared for delivery to the target areas and preserved at the target site. In addition, the enzyme must be able to gain access to the infecting bacteria, and be present in sufficient quantity to kill the growing bacteria. None of these methods would be useful in the treatment of gram negative bacteria, because the endolysins could not penetrate the outer membrane of such bacteria.
Attempts have been made to treat both gram-positive and gram-negative bacterial diseases of animals, but not plants, by use of lytic enzyme preparations extracted from bacteriophage infected bacteria or from bacteria expressing bacteriophage genes. WO 01/51073, WO 01/82945, WO 01/019385, US 2002/0187136 and US 2002/0127215 disclose methods of preventing and treating a variety of gram positive and gram negative bacterial infections of animals by use of lytic enzymes that may optionally include specific “holin lytic enzymes” or “holin enzymes”.
Since holins are not known to exhibit enzymatic function, and since examples of such holin lytic enzymes are not demonstrated or taught in WO 01/51073, WO 01/82945, WO 01/19385, US 2002/0187136 and US 2002/0127215, such enzymes appear to represent a theoretical and undemonstrated enzyme defined by reference to a desirable characteristic or property. As correctly stated elsewhere by the same inventors: “Holin has no enzymatic activity” (refer WO 01/90331, page 9 line 12). Lytic enzymes, which form the basis for the methods disclosed in all of these PCT publications, are internally defined: “The present invention is based upon the discovery that phage lytic enzymes specific for bacteria infected with a specific phage can effectively and efficiently break down the cell wall of the bacterium in question. At the same time, the substrate for the enzyme is not present in mammalian tissues.” (WO 01/51073 paragraph 3, page 4). “The lytic enzymes produced by bacterial phages are specific and effective for killing select bacteria.” (paragraph 2, page 7).
The term “holin enzyme” as used in Claim #3 of WO 01/51073 refers to the enzymes defined in Claim #1 as “the group consisting of lytic enzymes, modified lytic enzymes and combinations thereof.” Similar references in the claims of WO 01/82945, WO 01/019385 and US 2002/0187136 and US 2002/0127215 may be found. None of these patent applications disclose or claim the use of holin or other phage derived proteins that lack enzymatic activity in any manner, including the formulation of a compound or method of treatment of animal or plant diseases.
WO 02/102405 discloses a method of preventing food poisoning in animals by inclusion of a purified preparation consisting of specific lytic enzymes and optionally, specific lytic “holin enzymes”. Again, since holins are not known to exhibit enzymatic function, it is unclear as to what is taught or specified in the claims, other than a theoretical and undemonstrated enzyme defined by reference to a desirable characteristic or property.
It has been suggested that a specific endolysin from a bacteriophage that attacks a gram negative bacterial plant pathogen might be effective in providing resistance to that pathogen if the endolysin gene were cloned and expressed in plants (Ozawa et al., 2001). This suggestion is most unlikely, since endolysins other than T4 lysozyme are not known to penetrate bacterial membranes, and Gram-negative bacteria have a distinctive outer membrane, the LPS barrier, that provides a strong environmental barrier that is impermeable to most molecules.
It has been demonstrated that a gene from a bacteriophage infecting Ralstoma solanacearum encodes a lytic peptide that is capable of lysing several R. solanacearum strains (Ozawa et al. 2001). These authors suggested that this lytic peptide of undisclosed sequence might be used to enhance resistance against R. solanacearum in transgenic tobacco plants. However, there is no teaching or suggestion that this lytic peptide has bacteriocidal or bacteriostatic ability against any bacteria other than certain strains of R. solanacearum. Indeed, this evidently species-specific lytic peptide was expressed in E. coli without report of damage to the producing E. coli strains (Ozawa et al. 2001. This is not unexpected, since phage are highly specific for their bacterial host strains, and are normally limited in host range to a small subset of strains within a given host species. Methods are urgently needed to enhance resistance of plants against a broader range of pathogenic bacteria than a few strains of one pathogenic species.
In all previously published cases wherein phage genes are reported or suggested for use in a transgenic approach, the phage genes either encoded enzymes or, in one case, a highly species specific lytic peptide. In all previously published cases wherein phage preparations are incorporated, used or described, enzymes or enzyme preparations are involved. These enzymes must be purified, buffered, prepared for delivery to the target areas and preserved at the target site.
Thus, the prior art fails to teach or describe the identification or use of phage proteins with wide anti-microbial activity against Gram-negative bacteria. The prior art also fails to teach the use genes encoding phage proteins with wide anti-microbial activity against Gram-negative bacteria. In particular, the prior art fails to teach the use of phage proteins that are capable of destabilizing or permeabilizing the outer bacterial membrane (the bacterial lipopolysaccharide or LPS barrier) for the control of Gram negative bacterial infections of plants.